Diffusion Models
Overview
Diffusion models learn a data distribution by gradually corrupting examples with noise and training a denoiser to reverse that process. Major approaches include DDPMs and score-based SDEs, latent diffusion and diffusion Transformers, flow and consistency methods, and discrete masked or continuous-time Markov diffusion; conditioning and guidance adapt them to specific tasks. They are used for images, video, audio, language, molecules, and scientific forecasting, with evaluations ranging from ImageNet FID to language perplexity and domain-specific measures of accuracy and uncertainty. A central tension is how to make sampling faster and generation more controllable without sacrificing quality, calibration, or diversity, while theory and ongoing audits address generalization, memorization, and privacy.
Latest papers
1,278 papers90 papers in the last four weeks, down 44% on the four weeks before. 0.7% of all new papers.
Data-to-Energy Stochastic Dynamics
Lavida-O: Elastic Large Masked Diffusion Models for Unified Multimodal Understanding and Generation
Diff-Feat: Diffusion-Based Cross-Modal Feature Extraction for Multi-Label Classification
IS-Diff: Improving Diffusion-Based Inpainting with Better Initial Seed
ECHO: Ego-Centric modeling of Human-Object interactions
MeSS: City Mesh-Guided Outdoor Scene Generation with Cross-View Consistent Diffusion
EEGDM: Label-Efficient EEG Representation Learning with Generative Diffusion Model
Double-Diffusion: Balancing Speed, Accuracy, and Uncertainty in Probabilistic Forecasting for Urban Sensor Networks
Critically-Damped Higher-Order Langevin Dynamics for Generative Modeling
position" variable, and the auxiliary variables are referred to as velocity", acceleration", etc. In this sense, it is possible to critically damp" the dynamics. Critical damping has been successfully introduced in Critically-Damped Langevin Dynamics (CLD) and Critically-Damped Third-Order Langevin Dynamics (TOLD++), but has not yet been applied to dynamics of arbitrary order. The proposed methodology generalizes Higher-Order Langevin Dynamics (HOLD), a recent state-of-the-art diffusion method, by introducing the concept of critical damping from systems analysis. Similarly to TOLD++, this work proposes an optimal set of hyperparameters in the -dimensional case, where HOLD leaves these to be user defined. Additionally, this work provides closed-form solutions for the mean and covariance of the forward process that greatly simplify its implementation. Experiments are performed on the CIFAR-10 and CelebA-HQ datasets, and validated against the FID metric.